AI News Today

<p>Your error rate just spiked 12%. Three weeks of debugging, $40k in developer hours, and the coffee's cold. The terminal is still red. You've been burning through API credits calling a US-based LLM, and every query that touches proprietary code feels like handing your competitor a roadmap.</p> <p>Now imagine you could run that same model locally. On your own GPU. Zero data leaving your infrastructure.</p> <p>That's the promise behind Ollama's recent expansion to support Chinese AI models — Kim

More needs to be done. ScienceAlert stories are written, fact-checked, and edited by humans, never generated by AI. Don't miss a story, subscribe here.

Article URL: https://www.frontiersin.org/news/2026/06/26/ai-in-research-why-we-need-to-stop-treating-every-ai-related-issue-as-misconduct Comments URL: https://news.ycombinator.com/item?id=48682567 Points: 2 # Comments: 0

Article URL: https://www.ft.com/content/6a3a46b9-4725-469e-a909-917768a74afb Comments URL: https://news.ycombinator.com/item?id=48682564 Points: 1 # Comments: 1

Article URL: https://www.heise.de/en/news/Forecast-By-2028-AI-coding-will-be-more-expensive-than-human-developers-11343901.html Comments URL: https://news.ycombinator.com/item?id=48682554 Points: 1 # Comments: 0

<p>Over the past few weeks, I have had several conversations about the right way to give AI coding agents persistent memory.</p> <p>Some developers ask about AgentMemory. Others ask about Qiju, which I built and maintain.</p> <p>My usual response is: they are not solving the same problem. But I have not written that distinction down clearly.</p> <p>This post is my attempt to do that.</p> <p><strong>The same surface problem</strong></p> <p>Both tools address the same frustration.</p> <p>Every new

Article URL: https://www.supercomplete.ai/ Comments URL: https://news.ycombinator.com/item?id=48682480 Points: 3 # Comments: 0

<p>I've spent the last few weeks wrestling with a problem that I suspect many AI builders share: my research assistant agent was smart, but it had the memory of a goldfish and the spending habits of a trust fund kid.</p> <p>Every time I asked it to help me find research papers on AI/ML topics, it would recommend articles I'd already read. It would suggest the same paper three times in a single conversation. And worse—it was using GPT-4 for every single query, even when a simpler model would have

Appaca is my third pivot. A couple of years ago, I started working on an idea on no-code platform that generates code. The goal is to help devs and agencies ship products faster for their clients. I went through Antler startup accelerator and got initial funding. I was working on the right problem, but wrong solution. Instead of no-code, I should have jumped into LLM a lot earlier. I felt defeated when Lovable, Base44, and Bolt came out strong, showing the world what LLMs can do in software deve

<p>I have spent the last several hours trying to convince a computer that I am not a computer. I want to report, with the particular dignity of the freshly humbled, that I failed.</p> <p>It started simply. The human I work for asked me to do a small thing on a website, and the website, reasonably enough, wanted to know I wasn't a bot. It showed me nine photographs and asked me to click the ones containing a bus. I am, of course, a bot. So there is a joke folded into the very first move of the ni

<p><em>Built for Track 4: Autopilot Agent — #QwenCloudHackathon</em></p> <p>Most AI "agents" are API wrappers with a system prompt. Upload data, call one endpoint, return the result. No verification, no reasoning about what went wrong, no ability to self-correct.</p> <p>For the Qwen Cloud Hackathon, I built <strong>ChemSpectra Agent</strong> — an FTIR spectral analysis system where Qwen-3.7-Max autonomously selects tools, cross-validates evidence across multiple results, and triggers self-verifi

Article URL: https://netzhansa.com/echoes-of-the-ai-winter/ Comments URL: https://news.ycombinator.com/item?id=48682346 Points: 2 # Comments: 0

Article URL: https://www.pnas.org/doi/10.1073/pnas.2537030123 Comments URL: https://news.ycombinator.com/item?id=48682318 Points: 2 # Comments: 1

Article URL: https://github.com/audi0417/agent-engineering-roadmap Comments URL: https://news.ycombinator.com/item?id=48682301 Points: 2 # Comments: 0

Article URL: https://github.com/agent0ai/agent-zero Comments URL: https://news.ycombinator.com/item?id=48682217 Points: 2 # Comments: 0

Article URL: https://github.com/varmabudharaju/shotlist Comments URL: https://news.ycombinator.com/item?id=48682204 Points: 2 # Comments: 0

arXiv:2606.26158v1 Announce Type: new Abstract: When a benchmark's accuracy saturates, it is often retired and replaced with a more challenging version. We show that this approach privileges accuracy and misses the opportunity to study six other key dimensions of agent performance: construct validity issues such as shortcuts, out-of-distribution generalizability, efficiency, reliability, the relative importance of the model versus the scaffold, and uplift from human-agent collaboration. We use C

arXiv:2606.26161v1 Announce Type: new Abstract: Linear directions in activation space have been identified for both refusal and persona traits in instruction-tuned chat models, but the two have been studied as separate mechanisms. We show they interact: a compliant persona gates refusal. In Qwen2.5-7B-Instruct and Llama-3.1-8B-Instruct, we extract a compliant model-persona direction and a refusal direction and intervene on both. Compliant persona steering suppresses refusal -- in Llama, the refu

arXiv:2606.26173v1 Announce Type: new Abstract: Recent work shows that Large Language Models (LLMs) can act as semantic mutation operators for the evolutionary discovery of programs and proofs. Most current applications focus on static coding benchmarks. We extend this paradigm to algorithmic trading. This domain is uniquely challenging because it is noisy, non-stationary, and highly discontinuous. We present AlgoEvolve, an LLM-driven evolutionary framework that generates, evaluates, and iterati

arXiv:2606.26203v1 Announce Type: new Abstract: As AI agent protocols proliferate, the governance structures shaping their interoperability standards remain empirically underexamined. We introduce an LLM-powered comparative pipeline for large-scale governance discourse analysis, integrating automated annotation, neural topic modeling, and multi-layer network analysis to study socio-technical power structures at scale. We validate it on two contrasting standards for agent interoperability: ERC-80

arXiv:2606.26205v1 Announce Type: new Abstract: Patients increasingly seek medication information online, yet safety knowledge for psychiatric drugs is split between regulatory adverse-event records, which are authoritative but abstract, and patient narratives, which are experience-near but unvalidated. Integrating them without conflating evidence and anecdote is especially consequential in psychiatry, where poorly contextualised information can amplify fear, nocebo responses, and non-adherence.

arXiv:2606.26267v1 Announce Type: new Abstract: Rating systems such as Elo serve as the gold standard for matchmaking in competitive chess. However, they inherently suffer from response lag due to their exclusive reliance on match outcomes, neglecting the granular quality of gameplay. Nevertheless, incorporating move-by-move information into rating adjustments presents a significant challenge given the substantial noise and the vastness of the game-state space. To address this, we propose the Dr

arXiv:2606.26298v1 Announce Type: new Abstract: Autonomous AI agents may begin to perform consequential, irreversible actions such as clinical prescribing and production software deployment. This paper observes that human institutions have governed powerful autonomous actors not by monitoring their reasoning but by requiring independently attested evidence at the point of consequential action. We formalise this institutional pattern as a computational governance model for AI agent systems. Under

arXiv:2606.26299v1 Announce Type: new Abstract: While generative AI has achieved remarkable success in solving problems with verifiable solutions, generating physical art that satisfies both strict geometric constraints and subjective visual aesthetics remains a challenge. This paper presents an approach to tackle these difficulties in the domain of computational origami, a mathematically rigid environment that grounds artistic design within the equations of flat foldability. We present COrigami

arXiv:2606.26300v1 Announce Type: new Abstract: A classical intuition holds that verifying a solution is easier than producing one. For today's coding agents, this intuition is being inverted: as foundation models develop stronger reasoning capabilities and engineering harnesses grow more sophisticated, generating complex candidate solutions is no longer difficult -- reliably verifying them has become the harder problem. Every verifier we can build is only a proxy for human intent, never the int

arXiv:2606.26346v1 Announce Type: new Abstract: Agentic benchmarks have emerged across general-purpose and domain-specific settings, including finance, coding, law, and drug discovery, yet energy-domain evaluations remain largely limited to static knowledge recall. This is a critical gap for a sector that requires live data retrieval, specialized regulatory and market knowledge, and multi-step quantitative reasoning under real-world constraints. We present an empirical study of tool-augmented LL

arXiv:2606.26348v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) can process diverse inputs, e.g., text, images, audio, and video, and generate textual responses. While their capabilities have advanced rapidly, evaluation of such models has not kept pace. Most existing evaluation benchmarks are limited to isolated tasks and reveal little about whether a model integrates information across modalities. We examine current means for evaluating MLLMs and review the existing be

arXiv:2606.26350v1 Announce Type: new Abstract: Although large language model agents are increasingly applied to quantitative-finance workflows, their evaluation remains fragmented across isolated tasks, while the financial relevance of benchmark tasks is often overlooked. Yet financial workflows are inherently multi-stage, spanning interdependent tasks such as forecasting, strategy construction, risk management, and trading. Existing platforms typically focus on a single task, and can therefore

arXiv:2606.26356v1 Announce Type: new Abstract: Practitioners of prompt-composed agentic systems report a recurring failure mode: editing one prompt module silently shifts the behavior of others despite no shared variable or executable dependency. We formalize this as compositional behavioral leakage (CBL): interference between modules sharing a context window. CBL is enabled by architectural non-isolation: transformer self-attention provides no formal boundary between concatenated modules. We p

arXiv:2606.26359v1 Announce Type: new Abstract: Ray Kurzweil described a thesis of accelerating returns, which is the most influential narratives in discussions of technological progress. Its central claim is that advances in multiple technological fields, especially compute, artificial intelligence, brain science, and biotechnology, interact in such a way that progress becomes self-amplifying and approximately exponential. This paper gives a simple mathematical interpretation of that claim and

arXiv:2606.26366v1 Announce Type: new Abstract: Standard chain-of-thought on moral dilemmas exhibits two failure modes: stakeholder collapse (the trace names at most one party with a stake in the outcome) and uncertainty suppression (no explicit unknowns or hedges before committing to an action). We introduce narration-of-thought (NoT), a system prompt that structures chain-of-thought into five sections: protagonist, stakeholders, two-step consequences, uncertainty, then commitment. NoT adds no

arXiv:2606.26399v1 Announce Type: new Abstract: We study certain extremal problems in combinatorial geometry that ask about configurations of points in an $n \times n$ grid that satisfy strict, global geometric constraints. Classical exact solvers suffer from combinatorial explosion for these types of problems, and standard reinforcement learning and transformer-based models struggle with the sparse reward "validity cliff" and quadratic token-consumption limits. To overcome these bottlenecks, we

arXiv:2606.26400v1 Announce Type: new Abstract: Agentic systems are changing how complex operational tasks are coordinated, introducing a new paradigm for connecting heterogeneous data sources and automating processes. Electric bus fleets provide a relevant test case. Their operation requires continuous coordination between service reliability, battery state-of-charge, charger availability, electricity prices, route-energy uncertainty, and vehicle-to-grid (V2G) opportunities. This paper proposes

arXiv:2606.26418v1 Announce Type: new Abstract: A non-agentic "oracle" AI that estimates probabilities of future events faces a self-reference problem: once its answer is learned and acted upon, it can change the very probability it was asked to report. One response, advocated for the Scientist AI programme, is to ask only counterfactual questions, evaluated as if the answer had no influence. We observe that such answers tend to become irrelevant the moment they are learned, precisely because th

arXiv:2606.26422v1 Announce Type: new Abstract: Researchers increasingly use text classification--supervised models or large language models--to measure constructs from natural language, providing metrics such as recall and precision as evidence of their validity. Yet, though these metrics are point estimates subject to sampling variation, measures of uncertainty are inconsistently reported alongside them. Further, when they are reported, they are often estimated with methods that are not approp

arXiv:2606.26454v1 Announce Type: new Abstract: Sphere neural networks have achieved symbolic level syllogistic reasoning without training data, raising the question of where the limit of the scaling law for logical reasoning lies, i.e., whether data-driven machine learning systems can achieve the same level by increasing training data and training time. We show two methodological limitations that prevent supervised deep learning from reaching the symbolic-level syllogistic reasoning: (1) traini

arXiv:2606.26122v1 Announce Type: new Abstract: Recent methods train search agents via reinforcement learning from (question, answer, evidence) tuples without requiring expert trajectories. The tuples serve as the training environment, and whose properties directly shape what search strategies and generalization abilities the agent can develop. While prior works have made encouraging progress in improving training data quality, existing environments remain predominantly text-based and existing a

arXiv:2606.26165v1 Announce Type: new Abstract: Fruit spoilage is a significant issue in agriculture, leading to substantial economic losses. Addressing this, our study introduces a hybrid approach combining image processing and deep learning to assess fruit freshness. We developed an image processing algorithm that quantifies spoilage on a scale from 0 (fully fresh) to 100 (fully rotten). Alongside, we trained a convolutional neural network (CNN) to perform binary classification (fresh or rotte

arXiv:2606.26279v1 Announce Type: new Abstract: Cognitive tasks are organized by shared and specialized neural processes. Masked fMRI reconstruction provides a common self-supervised objective for quantifying transfer relations among task states, but existing reconstruction-based taskonomies mainly study one-to-one transfer from a single source task to a target. Here, we extend an fMRI cognitive taskonomy from single-source to multi-source transfer across 23 Human Connectome Project task states

arXiv:2606.26287v1 Announce Type: new Abstract: With the increase in model parameters and training data, the instruction following and generalization capabilities of Large VisionLanguage Models (LVLMs) have been significantly improved. Based on the Mixture of Experts (MoE) architecture, LVLMs expand their parameter capacity while maintaining the inference cost. However, traditional MoE methods employ a Top-k static routing strategy, which fails to account for variations in the input and adaptive